A Comparison of Empirical Ranking Methods of Frequency and Severity Ratings of Clinical Presentations

Abstract

This study compares five methods of ranking Likert ratings on frequency and severity scales. Data were drawn from an international online survey conducted as part of a practice analysis with 91 diplomates of the American Chiropractic Board of Radiology. A total of 129 clinical presentations had been rated on two scales. The frequency scale specifies how often each condition is seen in practice from 1 (never) to 5 (daily). The severity scale indicates how severe each condition is to the patient from 1 (no risk) to 5 (severe risk). These ratings were used in five methods of ranking to identify the 10 most important conditions reported by chiropractic radiologists. The similarity of ranks across the five methods was then analyzed with Spearman’s rank correlations. The results of the study indicate that the Rasch model provides the most precise results about the importance of conditions seen in practice.